ArticlePDF Available

Prediction of popcorn hybrid and composite means

Authors:

Abstract and Figures

The objectives of this study were to evaluate the combining ability of 10 popcorn populations in a circulantdiallel; to test the viability of superior hybrids; and to identify genetic composites for intrapopulation breeding. In fourcontrasting environments, 30 treatments were evaluated for grain yield (GY), plant height (PLH) and popping expansion(PE), in a random block design with four replications. Results indicated that the evaluation of treatments in a larger groupof environments favors the expression of variability in genotypes. The additive effects for PE and the dominance effects forGY and PLH were highest. GY and PE of the combinations UNB2U-C1 x Angela and Braskalb x Angela were outstanding.The predicted mean PE and GY were highest for hybrid UNB2U-C1 x Angela and the composite formed by these parents(26.54 mL g-1 and 1,446.09 kg ha-1 respectively).
Content may be subject to copyright.
287 Crop Breeding and Applied Biotechnology 7: 287-295, 2007
Prediction of popcorn hybrid and composite means
Crop Breeding and Applied Biotechnology 7: 287-295, 2007
Brazilian Society of Plant Breeding. Printed in Brazil
Received 03 July 2006
Accepted 12 May 2007
1 LMGV, CCTA, Universidade Estadual do Norte Fluminense Darcy Ribeiro, Av. Alberto Lamego, 2000, Parque Califórnia, 28.013-602, Campos dos Goytacazes-
RJ, Brasil. *E-mail: amaraljr@uenf.br
Prediction of popcorn hybrid and composite means
Ramon Macedo Rangel1, Antonio Teixeira do Amaral Júnior1*, Alexandre Pio Viana1, Silvério de Paiva Freitas Júnior1, and Messias
Gonzaga Pereira1
ABSTRACT - The objectives of this study were to evaluate the combining ability of 10 popcorn populations in a circulant
diallel; to test the viability of superior hybrids; and to identify genetic composites for intrapopulation breeding. In four
contrasting environments, 30 treatments were evaluated for grain yield (GY), plant height (PLH) and popping expansion
(PE), in a random block design with four replications. Results indicated that the evaluation of treatments in a larger group
of environments favors the expression of variability in genotypes. The additive effects for PE and the dominance effects for
GY and PLH were highest. GY and PE of the combinations UNB2U-C1 x Angela and Braskalb x Angela were outstanding.
The predicted mean PE and GY were highest for hybrid UNB2U-C1 x Angela and the composite formed by these parents
(26.54 mL g-1 and 1,446.09 kg ha-1 respectively).
Key words: Popcorn, circulant diallel, combining ability, popping expansion, composite estimates.
INTRODUCTION
Popcorn is a crop of high economic value and
consumption in Brazil is on the rise (Brugnera et al.
2003). For being cheap and tasty, always freshly popped
for clients and served hot, popcorn is very popular. For
many people that work informally, popcorn is a direct
source of family income. Until recently, the commercial
planting of popcorn in Brazil was rather modest and
significant grain importations, above all from the USA
and Argentina, were necessary (Galvão et al. 2000).
However, the market situation of the crop is changing.
According to information of the packaging industry,
grain importations have dropped markedly, mainly due
to the large-scale use of the modified single hybrid IAC-
112 (Sawazaki 2001) and of North American hybrids
cultivated in the country (Sawazaki et al. 2003).
In spite of the progress, the low number of
popcorn cultivars is still a constraint to further expansion
of the crop in the country (Scapim et al. 2002).
Consumers generally associate quality popcorn to
imported grain, which implies that good popcorn must
come from abroad. Sawazaki et al. (2000) and Galvão et
al. (2000) demonstrated that this premise is not true, in
an analysis of hybrids between the lines Guarani and
IAC-64, in the states of São Paulo and Minas Gerais.
The yield and popcorn quality were similar to the best
North American hybrids.
To establish superior genotypes, the diallel
analysis lends itself as viable alternative, even when
the number of parents involved is high. In this situation,
the circulant diallel is a particularly interesting
methodology, for requiring only one sample of all
possible parent combinations (Cruz et al. 2004).
Crop Breeding and Applied Biotechnology 7: 287-295, 2007 288
RM Rangel et al.
Veiga et al. (2000) demonstrated the efficiency of
circulant in comparison with complete diallels, to
classify parents as well as estimate parameters of general
and specific combining ability. Ferreira et al. (2004)
further verified the superior efficiency of the circulant
over the complete diallel regarding the effects of general
and specific combining ability. The authors evaluated
maize ear yield in a circulant diallel and stated that a
reduction of 30% in the number of crosses did not affect
the choice of the best parents and hybrids.
This study was carried out to evaluate the
combining ability of 10 popcorn populations in a
circulant diallel, to check the feasibility of establishing
commercial hybrids and identify genetic composites for
intrapopulation breeding.
MATERIAL AND METHODS
Diallel crosses in a circulant scheme were
performed between the populations Beija-flor, Branco-
Viçosa, Viçosa-Viçosa, SE013-Maringá, PA038-Maringá,
BRS Angela, Braskalb, Viçosa-UENF, UNB2U-C1 and
UNB2U-C2, based on the algorithm of Kempthorne and
Curnow (1961).
To create the hybrid combinations, the 10
populations were grown in 6.00 m long rows (row
spacing 1.00 m and within-row plant spacing 0.40 m),
and paired in all necessary combinations, in March 2003.
The plants of the pairs of rows were crossed by hand.
For this purpose, the mature tassels of a particular plant
of a row were wrapped in kraft paper bag. Then the bag
was used to cover an ear ready for pollination of a plant
of the corresponding row-pair. Approximately 100 ears
per row-pair were pollinated to generate a sufficient seed
quantity for the follow-up procedures.
The diallel crosses were based on the algorithm
proposed by Kempthorne and Curnow (1961), where: p=
number of parents under study; s= number of hybrid
combinations of each parent, s < p-1 and s > 3 for diallels
that include only F1s and s > 2 when the parents are also
included; ps/2= total number of crosses; and
K=(p+1-s)/2: whole number. Three crosses of each
population with the others were used for this purpose (s =3).
The hybrids, parents and controls (IAC-112,
Produtor-Maringá, PR023-Maringá, UNB2U-C0 and
Viçosa) were evaluated in two contrasting
environments, in the State of Rio de Janeiro: Colégio
Estadual Agrícola Antônio Sarlo, in Campos dos
Goytacazes, at the Experimental Station of PESAGRO-
RIO, Itaocara, planted on 09/12/2004 and 14/12/2004,
respectively. In each environment, 30 treatments,
corresponding to the 10 pre-selected parents, 15 hybrids
and 5 controls were arranged in random blocks with
four replications, planted in single rows of 10.00 m length,
(row spacing 1.00 m and within-row plant spacing 0.20 m).
The following traits were evaluated: grain yield (GY),
in kg ha-1, measured by weighing the grains after shelling;
plant height (PLH), in cm, measured after tasseling as the
distance between the soil level to the insertion height of
the flag leaf of 10 healthy plants; and popping expansion
(PE), in mL g-1. For this purpose, two 30g samples of each
plot were popped for 2 minutes and 30 seconds at 270 ºC
in an expansion measuring device developed by
EMBRAPA-CNPDIA.
The individual and joint diallel analyses were carried
out using the software package GENES (Cruz 2006) and
the genetic-statistical model Yij = µ + gi + gj + sij + ξij and
Yij = µ + gi + gj + sij + lk + glik + gljk + slijk + ξij, where:
Yij = mean of the observation associated to the hybrid
combination ij (i =/ j) or to the ith parent (i = j); µ = general
constant; gi and gj = effects of the general combining
ability associated to the ith and jth parent, respectively;
sij = effect of the Specific combining ability between the
ith and jth parent; lk = effect of the kth environment or site;
glik and gljk = effects of the interaction between the general
combining ability associated to the ith and jth parent and
the kth environment, respectively; slijk = effect of the
interaction between the specific combining ability between
the ith and jth parent and the kth environment; and ξij. =
mean experimental error.
The estimators of the components of the general
and specific combining ability, respectively φgi and φsi,
were expressed as: φgi = , and φsi =
, where: MSGA= mean square of the
general combining ability; MSGA= mean square of the
specific combining ability; MSE = mean square of the
error; b = number of blocks; s = number of crosses; and
p = number of parents. In the joint analysis, the
estimators of the components of the general and specific
combining ability in interaction with the environment
were expressed by:
φgiA= and φsiA = ,
289 Crop Breeding and Applied Biotechnology 7: 287-295, 2007
Prediction of popcorn hybrid and composite means
where: MSGCAxE and MSCSAxE = general and specific
combining ability in interaction with the environment,
respectively.
The number of composites was estimated by the
expression: n = Cx
p , where x is the number of parents
involved in the composite and p the number of parents
involved in the diallel (Cruz 2006). The means of the
biparental composites were estimated by the expression
PV = , where PV = Yij predicted value; and
Yij = parent mean, when i = j, or of the hybrid, when
i =/ j, of the selected variable. The composites with
three parents were estimated by the
PV = .
RESULTS AND DISCUSSION
The analysis of variance for the combining ability
of the environment in Campos dos Goytacazes
(Table 1) indicated sufficient genetic variability in the
parents and hybrids of the diallel, due to the genetic
additive and non-additive effects.
The estimates of the quadratic components due to
the GCA effects revealed superiority of the genetic additive
over the non-additive effects for trait PE (12.6032). For GY
and PLH the non-additive effects were higher. The
predominance of genetic additive effects in popcorn for
PE had also been described by Pacheco et al. (1998), Pereira
and Amaral Júnior (2001) and Simon et al. (2004). With
respect to GY, Andrade et al. (2002), Pereira and Amaral
Júnior (2001) and Simon et al. (2004) also demonstrated
the greater importance of non-additive effects.
The GCA effects (g^i) for the environment of Campos
dos Goytacazes expressed positive GY estimates for the
SV df Mean squares1
GY PLH PE
Genotypes 24 596481.7708 ** 283.1197 ** 84.9479 **
GCA 9628353.0421 ** 340.3592 ** 204.6025 **
SCA 15 577359.0084 ** 248.7750 ** 13.1555 **
Error 87 45880.4597 54.3960 2.9512
Mean square effects
GCA 36404.5363 17.8726 12.6032
SCA 132869.6371 48.5947 2.5510
Table 1. Values and significances of the mean squares of popcorn genotypes, partitioned in general and specific combining ability as well
as the mean square effects of the combining ability for three traits evaluated in a circulant diallel - Campos dos Goytacazes, RJ
** = Significant at 1% probability by the F test; * = Significant at 5% probability by the F test
1/ GY = grain yield in Kg ha-1; PLH = plant height in cm; and PE = grain popping expansion in mL g-1
parents UNB2U-C1, Viçosa-UENF, Angela, UNB2U-C2
and Viçosa-Viçosa, suggesting the suitability of these
genotypes for breeding programs focused on the
improvement of weight and grain yield (Table 2).
For plant height (PLH), the g^i estimates of the
parents PA038-Maringá, SE013-Maringá and Viçosa-
Viçosa were positive, for PE, the positive g^i magnitudes
of the parents Angela, UNB2U-C1 and UNB2U-C2
were highest (6.1486; 2.5795; and 1.3271, respectively)
(Table 2).
Due to the positive g^i values associated to the
traits GY and PE, the parents UNB2U-C1, Angela and
UNB2U-C2 can be included in intrapopulation breeding
programs, with a view to the establishment of generations
with superior grain yield and popping expansion.
When evaluating the same genotypes in Campos
dos Goytacazes in the growing season 2003/2004,
Freitas Júnior et al. (2006) confirmed the recommendation
of UNB2U-C1, Angela and UNB2U-C2, besides Viçosa-
Viçosa, for the development of superior progenies in
intrapopulation breeding programs.
With respect to the s^ij estimates (Table 2) 10
combinations were promising regarding GY, above all
the hybrids UNB2U-C1 x Angela and Viçosa-UENF x
UNB2U-C2 (estimates of 17.25 and 143.68, respectively).
These were not the highest s^ij , values, but were the
only pairs derived from parents with positive g^i
estimates and are therefore considered superior when
breeding for higher grain yields. The s^ij values alone do
not identify the best combinations, but must be
complemented by the evaluation of the g^i values, which
must be of favorable effect for at least one of the parents
(Cruz et al. 2004). The above hybrids obtained GY means
of 1511.25 and 1608.75 kg ha-1, respectively.
Crop Breeding and Applied Biotechnology 7: 287-295, 2007 290
RM Rangel et al.
Table 2. Estimates of phenotypic means and effects of the general (g^
i) and specific combining ability (s^ij) for three traits1 evaluated in
a circulant diallel
Traits Estimates of g^i effects
1 2 3 4 5 6 7 8 9 10
GY 269.00 -210.08 98.35 -69.29 -221.11 21.99 -142.64 163.71 -23.27 113.34
PLH -0.18 -5.00 0.04 1.21 -1.37 -0.90 0.69 -4.78 4.92 5.36
PE 2.57 -1.66 -1.23 -2.68 -1.74 6.14 -0.52 1.32 -1.86 -0.32
Traits Estimates of the S^
ij effects (above the diagonal) of the phenotypic means of the parents (along the
diagonal) and of the hybrids (below the diagonal)
1 2 3 4 5 6 7 8 9 10
GY 1703.75 96.61 17.25 -39.35
1PLH 157.87 -3.26 0.51 0.66
PE 24.20 2.59 -0.87 -2.07
GY 563.75 553.83 -99.02 -16.63
2PLH 188.50 5.57 10.05 2.53
PE 13.92 2.77 -0.65 1.11
GY 1038.75 181.29 143.68 396.92
3PLH 150.62 -1.94 -0.45 15.70
PE 15.32 1.79 1.16 -0.83
GY 436.25 466.32 179.56 610.44
4PLH 153.75 4.74 5.15 1.84
PE 12.85 0.01 0.43 0.86
GY 1347.50 635.00 231.39 -76.47
5PLH 152.37 152.62 8.99 -2.06
PE 22.30 14.45 -0.87 0.13
GY 1511.25 1568.75 1063.75 -204.59
6PLH 156.62 153.50 149.00 10.08
PE 26.72 26.12 31.95 -3.46
7GY 1290.00 751.25 1340.00 896.25
PLH 158.37 158.37 156.00 156.50
PE 18.85 16.02 18.90 18.27
8GY 1140.00 1608.75 1763.75 1233.75
PLH 147.62 152.00 158.37 145.37
PE 19.65 20.12 17.52 20.37
GY 1675.00 1290.00 1190.00 752.50
9PLH 177.87 168.50 169.75 152.12
PE 14.92 14.75 14.37 15.77
GY 1857.50 1018.75 1133.75 1265.00
10 PLH 165.62 159.12 171.75 163.00
PE 16.72 16.92 21.22 19.45
G
e
n
o
t
y
p
e
s
1 GY = grain yield in Kg ha-1; PLH = plant height in cm; and PE = grain popping expansion in mL g-1
1 = UNB2U-C1; 2 = Braskalb; 3 = Viçosa-UENF; 4 = PA038-Maringá; 5 = Branco-Viçosa; 6 = Angela; 7 = Beija-flor; 8 = UNB2U-C2; 9 = SE013-
Maringá; and 10 = Viçosa-Viçosa
291 Crop Breeding and Applied Biotechnology 7: 287-295, 2007
Prediction of popcorn hybrid and composite means
predominance of the additive genetic effects on PE, due
to the non-significance of SCA at 5% probability, while
GY and PLH were significant at a level of 1% for SCA.
For GY, the effects for the general combining
ability of the parents UNB2U-C1, Viçosa-UENF, Angela,
UNB2U-C2, and Viçosa-Viçosa were positive (Table 4),
indicating these as genotypes of interest when focusing
on higher grain yields.
In relation to PE, the g^i values of the parents
Angela, UNB2U-C1, UNB2U-C2, Beija-flor, and Viçosa-
Viçosa were positive (6.2514; 1.9294; 1.8021; 1.0850; and
0.2759) (Table 4). The g^i estimates of the parents Angela,
UNB2U-C1, UNB2U-C2 and Viçosa-Viçosa were also
positive for grain yield. It is noteworthy that, with
exception of parent Viçosa-Viçosa, these genotypes
were the ones recommended for intrapopulation
breeding in the environment of Campos dos Goytacazes
(Table 2), i.e., a single breeding program can be
implemented for the environments studied.
The GY values were positive for all combinations,
with exception of Branco-Viçosa x Viçosa-Viçosa, which
had a negative value and is therefore uninteresting
(Table 4). Especially interesting were the hybrids
UNB2U-C1 x Angela, Viçosa-UENF x UNB2U-C2 and
Angela x Viçosa-Viçosa, with s^ij values of 542.34, 415.68,
and 175.17, respectively, besides positive g^i estimates.
It is worth pointing out that the first, third and fourth
highest phenotypic means for GY were found for the
hybrids UNB2U-C1 x Angela, Viçosa-UENF x UNB2U-
C2 and Angela x Viçosa-Viçosa, which confirms that they
are in fact superior.
With respect to PLH, the combinations Viçosa-
UENF x Beija-flor, PA038-Maringá x UNB2U-C2 and
Branco-Viçosa x Viçosa-Viçosa are indicated for
For the trait plant height (PLH), owing to the
strong winds that generally blow in Campos dos
Goytacazes, combinations with negative and high s^ij
values are ideal. From this viewpoint, the combinations
UNB2U-C1 x Branco-Viçosa, Viçosa-UENF x Beija-flor,
Viçosa-UENF x UNB2U-C2 and Branco-Viçosa x Viçosa-
Viçosa were superior, above all the first pair, with
negative g^i values in both parents and a phenotypic
value of 152.37 cm.
For mean popping expansion (PE), the
combinations UNB2U-C1 x Angela and Braskalb x
Angela, in this order, performed best in the environment
of Campos dos Goytacazes (26.72 mL g-1 and 26.12 mL
g-1, respectively). Angela is one of the parents in both
combinations. Considering that Angela was also the
parent with the highest positive g^i value, the absence
of the pair UNB2U-C1 x Angela among the superior
hybrids, based on s^ij is evidence of the need to select
the most promising hybrids more accurately.
For Ferrão et al. (1985), this absence is justified
when g^i and g^j estimates are highly discrepant and,
consequently, recommend that an indication of superior
hybrids must be based on estimates of the phenotypic
means. It was concluded that the best hybrids for PE in
Campos dos Goytacazes were UNB2U-C1 x Angela and
Braskalb x Angela.
In the analysis of the set of evaluated traits, the
hybrids UNB2U-C1 x Angela and Braskalb x Angela
outmatched the other combinations in the growing
season 2004/2005. For Freitas Júnior et al. (2006), these
hybrids were also superior in the growing season 2003/
2004; Braskalb x Angela stood out particularly for PE.
The analysis of variance for combining ability for
the environment of Itaocara (Table 3), ratified the
SV df Mean squares1
GY PLH PE
Genotypes 24 1468965.4583 ** 422.6479 ** 117.1242 **
GCA 91202721.5332 ** 452.9005 ** 305.5664 **
SCA 15 1628711.8018 ** 404.4900 ** 4.0588 ns
Error 87 221856.2045 67.1904 3.6667
Mean square effects
GCA 61304.0830 24.1068 18.8687
SCA 351713.8993 84.3248 0.0980
Table 3. Values and significances of the mean squares of popcorn genotypes, partitioned in general and specific combining ability, as well
as the mean square effects of the combining ability for three traits evaluated in a circulant diallel - Itaocara, RJ
** = Significant at 1% probability by the F test; * = Significant at 5% probability by the F test
1 GY = grain yield in Kg ha-1; PLH = plant height in cm; and PE = popping expansion in mL g-1
Crop Breeding and Applied Biotechnology 7: 287-295, 2007 292
RM Rangel et al.
Traits Estimates of g^i effects
1 2 3 4 5 6 7 8 9 10
GY 171.10 -277.30 95.14 -326.59 -222.43 153.84 -40.17 248.96 -102.08 299.52
PLH 2.01 -9.21 -0.98 -2.42 -4.54 0.49 3.52 1.16 4.08 5.87
PE 1.92 -0.69 -1.20 -5.01 -0.24 6.25 1.08 1.80 -4.17 0.27
Traits
Estimates of the S^ ij effects (above the diagonal) of the phenotypic means of the parents (along the
diagonal) and of the hybrids (below the diagonal)
1 2 3 4 5 6 7 8 9 10
GY 1703.75 316.12 542.34 351.36
1PLH 209.50 1.75 14.21 5.13
PE 24.20 1.44 -1.02 -0.01
GY 957.50 148.25 384.77 375.64
2PLH 188.50 5.57 10.05 2.53
PE 19.75 0.87 -0.28 -0.97
GY 1611.25 173.57 415.68 501.73
3PLH 208.62 -1.06 5.92 6.00
PE 19.15 0.52 0.26 -2.03
GY 476.25 189.92 650.98 833.11
4PLH 202.87 -6.75 10.94 12.40
PE 11.40 -0.93 -0.07 0.04
GY 2231.25 1123.75 564.32 -84.79
5PLH 215.25 207.12 5.94 -8.09
PE 24.07 20.45 -0.39 -1.04
GY 2833.75 1991.25 1841.25 175.17
6PLH 232.75 212.87 205.00 4.23
PE 28.10 27.37 33.15 0.75
GY 2448.75 2033.75 2195.00 1431.25
7PLH 226.70 220.37 217.50 216.00
PE 23.95 21.05 21.35 23.00
GY 2313.75 2726.25 2078.75 1973.75
8PLH 210.50 222.12 208.00 217.50
PE 21.07 21.80 16.80 25.37
GY 2461.25 2188.75 2206.25 903.75
9PLH 225.12 228.62 221.50 212.75
PE 13.52 11.67 16.12 13.85
GY 1958.75 2595.00 2103.75
10 PLH 209.25 226.62 223.50
PE 19.92 28.22 21.62
Table 4. Estimates of the phenotypic means and effects of the general (g^
i) and specific combining ability (s^ij) for three traits1/ evaluated
in a circulant diallel
G
e
n
o
t
y
p
e
s
1 GY = grain yield in Kg ha-1; PLH = plant height in cm; and PE = grain popping expansion in mL g-1
1 = UNB2U-C1; 2 = Braskalb; 3 = Viçosa-UENF; 4 = PA038-Maringá; 5 = Branco-Viçosa; 6 = Angela; 7 = Beija-flor; 8 = UNB2U-C2; 9 = SE013-
Maringá; and 10 = Viçosa-Viçosa
293 Crop Breeding and Applied Biotechnology 7: 287-295, 2007
Prediction of popcorn hybrid and composite means
environments favored the expression of variability
among the genotypes for the evaluated traits, as a result
of the effects of combining ability.
Freitas Júnior et al. (2006) observed no significant
effect of the environment on PE in an evaluation of the
diallel components in the growing season 2003/2004. In
our study however, the evaluation in trials in four
environments, in the growing seasons 2003/2004 and
2004/2005, detected significance of the environmental
effect (1% probability). This corroborates the idea that
a larger number of environments is favorable for the
identification of variability in the genotypes.
For grain yield, the significance at 1% probability
by the F test for GCA x environment and SCA x
environment interaction (Table 5) allows the conclusion
that parents as well as hybrids revealed differences in
the evaluated environments. However, the mean-square
analysis of the effects showed a difference between
the GCAxE and SCAxE values (8130.4869 and
43545.0214, respectively) which indicates a greater
differentiation in the hybrids than the parents.
A significance at 1% probability by the F test
regarding PE, for genotypes x environment, can be
associated to the significance at 1% probability for the
GCA x environment interaction. Although the values
for GCAxE and SCAxE are equal when observing the
mean square effects, it is assumed that genes with
additive effect prevail in the trait expression, based on
breeding for reduced plant size, since these are the only
pairs with negative S^ij estimates that were derived from
at least one parent with negative g^i value.
For PE, the S^ij values were positive for the hybrids
UNB2U-C1 x Branco-Viçosa, Braskalb x Angela, Viçosa-
UENF x Beija-flor, Viçosa-UENF x UNB2U-C2, PA038-
Maringá x Viçosa-Viçosa, and Angela x Viçosa-Viçosa.
The best one was Angela x Viçosa-Viçosa, due to the
parents with positive g^i estimates. However, the highest
estimates of phenotypic means for PE were observed,
hierarchically, for the hybrids: Angela x Viçosa-Viçosa,
UNB2U-C1 x Angela and Braskalb x Angela. According
to Ferrão et al. (1985), this can be explained by the
deviations between g^i and g^j , that were discrepant
enough to attain a negative S^ij value for the pair UNB2U-
C1 x Angela, which indicates the cited hybrids based
only on the mean. The best combinations were therefore
UNB2U-C1 x Angela, Braskalb x Angela and Angela x
Viçosa-Viçosa.
For the environment of Itaocara, the best hybrids
were UNB2U-C1 x Angela, Braskalb x Angela and Angela
x Viçosa-Viçosa. The second stood out mainly for
popping expansion and the others for grain yield as
well.
The significance for the combining abilities
revealed the existence of variability due to the additive
and non-additive genetic effects (Table 5). It was
therefore inferred that the evaluation in four
SV df Mean squares1
GY PLH PE
Genotypes 24 1879863.7656 ** 924.0285 ** 302.2309 **
GCA 91773191.8931 ** 1109.7318 ** 777.9206 **
SCA 15 1943866.8891 ** 812.6066 ** 16.8170 **
Environments (E) 327613517.7291 ** 70649.3538 ** 116.9652 **
Genotypes x E 72 259986.1753 ** 134.4705 ns 12.8757 **
GCA x E 27 232428.4915 ** 104.0195 ns 20.0224 **
SCA x E 45 276520.7856 ** 152.7411 ns 8.5877 ns
Error 348 102340.7000 107.0912 5.4761
Mean squares of the effects
GCA 26107.0498 15.6662 12.0694
SCA 115095.3868 44.0947 0.7088
GCAxE 8130.4869 -0.1919 0.9091
SCAxE 43545.0214 11.4124 0.7779
Table 5. Values and significances of the mean squares of popcorn genotypes, general and specific combining ability, and their interactions
with the environment, as well as the mean square effects of the combining ability for three traits evaluated in a circulant diallel in four
environments
** = Significant at 1% probability by the F test; * = Significant at 5% probability by the F test
1 GY = grain yield in Kg ha-1; PLH = plant height in cm; and PE = grain popping expansion in mL g-1
Crop Breeding and Applied Biotechnology 7: 287-295, 2007 294
RM Rangel et al.
the GCA and SCA values (12.0694 and 0.7088,
respectively).
The mean square analysis of the effects for PE
shows that the best option for the use of the parents in
intrapopulation breeding programs is by the formation
of genetic composites. For GY and PLH, breeding for
superior genotypes requires hybridizations, based on
the manifestation of the heterotic effect of these
combinations (Cruz et al. 2004).
Of the predicted composites, the best GY and PE
expressions were found for the biparental composite,
in particular the one derived from the parents UNB2U-
C1 and Angela. This combination performed best in all
four evaluated environments (1446.09 kg ha-1 and 26.54
mL g-1).
ACKNOWLEDGEMENTS
This research was funded by the CNPq and
FAPERJ.
Predição de médias de híbridos e compostos de milho
pipoca
RESUMO - O intento deste trabalho foi avaliar a capacidade combinatória de dez populações de milho pipoca, por meio de
dialelo circulante; averiguar a viabilidade de obtenção de híbridos superiores; e identificar compostos genéticos para o
melhoramento intrapopulacional. Trinta tratamentos foram avaliados para rendimento de grãos (RG), altura de planta
(ALTP) e capacidade de expansão (CE), em quatro ambientes contrastantes, utilizando o delineamento em blocos ao acaso
com quatro repetições. Os resultados indicaram que a avaliação dos tratamentos em um grupo maior de ambientes favoreceu
a expressão da variabilidade entre genótipos. Houve superioridade dos efeitos aditivos para CE e de dominância para RG e
ALTP. As combinações UNB2U-C1 x Angela e Braskalb x Angela destacaram-se para RG e CE. O híbrido UNB2U-C1 x
Angela e o composto formado por estes genitores revelaram-se superiores, simultaneamente, para CE e RG, com respectivas
médias preditas de 26,54 mL g-1 e 1.446,09 kg ha-1.
Palavras-chave: milho pipoca, dialelo circulante, capacidade combinatória, capacidade de expansão, predição de composto.
REFERENCES
Andrade RA, Cruz CD, Scapim CA, Silvério L, Pinto RJB and
Tonet A (2002) Análise dialélica da capacidade combinatória
de variedades de milho pipoca. Acta Scientiarum 24: 1197-
1204.
Brugnera A, Von Pinho RG, Pacheco CAP and Alvarez CGD
(2003) Resposta de cultivares de milho pipoca a doses de
adubação de semeadura. Revista Ceres 50: 417-429.
Cruz CD (2006) Programa Genes: versão Windows:
aplicativo computacional em genética e estatística.
Editora UFV, Viçosa, 648p.
Cruz CD, Regazzi AJ and Carneiro PCS (2004) Modelos
biométricos aplicados ao melhoramento genético. 3rd
ed. Editora UFV, Viçosa, 480p.
Ferreira FM, Ribeiro Júnior JI, Pacheco CAP, Silva CHO and
Cruz CD (2004) Efficiency of circulant diallels as compared
to complete diallels for the estimation of combining ability.
Crop Breeding and Applied Biotechnology 4: 145-151.
Ferrão RG, Silva JC and Cruz CD (1985) Avaliação da capacida-
de combinatória de oito linhagens de milho em um sistema
dialélico desbalanceado. Revista Ceres 32: 283-292.
Freitas Júnior SP, Amaral Júnior AT, Pereira MG, Cruz CD and
Scapim CA (2006) Capacidade combinatória em milho pipo-
ca por meio de dialelo circulante. Pesquisa Agropecuária
Brasileira 41: 1599-1607.
Galvão JCC, Sawazaki E and Miranda GV (2000) Comporta-
mento de híbridos de milho pipoca em Coimbra, Minas Ge-
rais. Revista Ceres 47: 201-218.
Kempthorne O and Curnow RN (1961) The partial diallel cross.
Biometrics 17: 229-250.
Pacheco CAP, Gama EP, Guimarães PEO, Santos MX and
Ferreira AS (1998) Estimativas de parâmetros genéticos nas
populações CMS-42 e CMS-43 de milho pipoca. Pesquisa
Agropecuária Brasileira 33: 1995-2001.
Pereira MG and Amaral Júnior AT (2001) Estimation of genetic
components in popcorn based on the nested design. Crop
Breeding and Applied Biotechnology 1: 3-10.
Sawazaki E, Paterniani MEAGZ, Castro JL, Gallo PB, Galvão
JCC and Saes LA (2000) Potencial de linhagens de popula-
ções locais de milho pipoca para síntese de híbridos.
Bragantia 59: 143-151.
Sawazaki E (2001) A cultura do milho pipoca no Brasil. O
Agronômico 53: 11-13.
295 Crop Breeding and Applied Biotechnology 7: 287-295, 2007
Prediction of popcorn hybrid and composite means
Sawazaki E, Fantin GM, Dudienas C and Castro G (2003) Resis-
tência de genótipos de milho pipoca a doenças. Revista de
Agricultura 78: 149-164.
Scapim CA, Pacheco CAP, Tonet A, Braccini AL and Pinto RJB
(2002) Análise dialélica e heterose de populações de milho
pipoca. Bragantia 61: 219-230.
Simon GA, Scapim CA, Pacheco CAP, Pinto RJB, Braccini AL
and Tonet A (2004) Depressão por endogamia em popula-
ções de milho pipoca. Bragantia 63: 55-62.
Veiga RD, Ferreira DF and Ramalho MAP (2000) Eficiência
dos dialelos circulantes na escolha de genitores. Pesquisa
Agropecuária Brasileira 35: 1395-1406.
... As sete cultivares de milho-pipoca são: zélia (híbrido triplo), Jade (híbrido triplo), iAC 112 (híbrido simples modificado), IAC 125 (híbrido triplo), BRS Angela (variedade), RS-20 (variedade) e UFVM2-Barão-Viçosa (variedade). Verificase , portanto, que apesar dos avanços o número de variedades e híbridos de linhagens comerciais de milho-pipoca, é reduzido em relação à crescente demanda pelo produto (VENDRUSCOLO et al., 2001; MATTA; ViANA, 2001; ANDRADE et al., 2002; DAROS et al., 2004; FREiTAS JúNiOR et al., 2006; SCAPiM et al., 2006; RANGEL et al., 2007; RANGEL et al., 2008; SANTOS et al., 2008; ViEiRA et al., 2009). Na Região Noroeste do Paraná existem empacotadoras que comprariam o produto se os agricultores começassem a semear em larga escala (PiNTO et al., 2007). ...
... Em contrapartida, os resultados aqui encontrados foram semelhantes aos obtidos por Galvão, Sawazaki e Miranda (2000) e Sawazaki et al. (2000), os quais verificaram bons resultados de CE em São Paulo e Minas Gerais. Na avaliação de dialelo circulante para identificação de compostos a serem utilizados em programa de seleção recorrente intrapopulacional, Rangel et al. (2007) constataram que o melhor composto foi constituído pela reunião das populações UNB-2UC1 x BRS Angela com valores de capacidade de expansão (CE) e rendimento de grãos (RG) de 26,54 mL g -1 e 1.446,09 kg ha -1 , respectivamente. 1,UEM 2 e UEM 7, quanto ao RG entre os aludidos locais. ...
Full-text available
Article
Although advances are obtained, the number of varieties and hybrids from inbred lines of popcorn are reduced. The objective of this work was to evaluate new composites developed in Brazil and to indicate those which present better potential to initiate an intrapopulation breeding program. Trials were carried out during 2005/2006 cropping season. Treatments included 21 genotypes (hybrid from inbred lines, composites and synthetic) cultivated in two locations (Umuarama and Cidade Gaúcha), both located in Northwest of Paraná State. Grain yield (GY) and popping expansion (PE) were evaluated by individual and joint analyses of variance. Composites UEM 5, UEM 6 and UEM 7 obtained the highest quality of the PE, with similar values to the check treatments in both locations. GY of genotypes ranged more than popping expansion over locations. Composites UEM 5, UEM 6 and UEM 7 revealed satisfactory potential both PE and GY when evaluated in Umuarama, Paraná State, Brazil.Apesar dos avanços, o número de variedades e híbridos comerciais de linhagens de milho-pipoca é reduzido. Assim, o objetivo deste trabalho foi avaliar novos compostos desenvolvidos no Brasil e indicar os de maior potencial genético para iniciar um trabalho de melhoramento intrapopulacional. A pesquisa foi desenvolvida no ano agrícola 2005/2006. Os tratamentos constaram de 21 genótipos (híbridos de linhagens, compostos e sintéticos) cultivados em dois municípios, Umuarama e Cidade Gaúcha, ambos no Noroeste do Estado do Paraná. As características avaliadas por meio de análises de variância individuais e conjunta foram rendimento de grãos (RG) e capacidade de expansão (CE). Os compostos UEM 5, UEM 6 e UEM 7 expressaram alta qualidade da pipoca (CE), com valores semelhantes às testemunhas em ambos os locais. O rendimento dos genótipos foi mais instável do que a capacidade de expansão entre os locais avaliados. Os compostos UEM 5, UEM 6 e UEM 7 revelaram potencial satisfatório tanto para CE quanto para RG quando avaliados em Umuarama, PR.
Full-text available
Article
The quantification of genetic components in a base population is essential for the definition of the best breeding strategy to maximize the genetic gains. To evaluate several agronomic traits such as grain yield (GY) and popping expansion (PE) in half-sib and full-sib families derived from an open pollinated popcorn population, this study used a nested design, in two contrasting environments in the Northern Fluminense region of Rio de Janeiro, Brazil. Results showed: a) high significance of male and female within male sources of variation for most traits studied; b) absence of genotype by environmental interaction for PE; c) overdominance for GY and additivity for PE; d) low narrow sense of heritability for GY as opposed to PE; e) intermediate family heritability mean for GY and high for PE; and f) high genetic gain estimates for both GY and PE provided by the full-sib recurrent selection. These results indicate the perspective of success for intrapopulation breeding methods in popcorn for popping expansion allowing the prediction of satisfactory genetic gain for the Northern Fluminense.
Full-text available
Article
Com os objetivos de elevar a freqüência de alelos favoráveis e estimar os parâmetros genéticos e fenotípicos de algumas características agronômicas, as populações de milho pipoca CMS-42 e CMS-43 foram submetidas ao 2 0 ciclo de seleção entre e dentro de progênies de meios irmãos. Avaliaram-se 196 progênies de cada população, em látice simples 14 x 14, tendo como testemunhas os respectivos ciclos anteriores. O espaçamento utilizado foi de 1,0 x 0,20, para parcelas de 5,0 m com 25 plantas. Embora se tenha dado maior importância à capacidade de expansão (CE), as médias das prog6nies selecionadas evidenciaram a possibilidade de obtenção de materiais superiores também quan-to à produtividade. A CMS-43, de maneira geral, apresentou maior variabilidade que a CMS-42, refletida num maior componente da variáncia genética aditiva. As médias da CE na testemunha, no ensaio e nas progénies selecionadas foram: 14,62, 15,21 e 19,65 para a CMS-42, e 16,07, 16,85 e 22,31 para a CMS-43, respcctivamente. Os coeficientes de herdabilidade no sentido restrito em progênies de meios irmãos (h 2), quanto à CE, foram da ordem de 60%, com uma expectativa de ganho com a seleção entre progênies de 16,80% para a CMS-42, e de 19,12% para a CMS-43. Termos para indexação: Zea mays, capacidade de expansão, melhoramento intrapopulacional, herdabilidade, ganho genético. GENETIC PARAMETERS ESTIMATIVES ff4 CMS-42 AND CMS-43 POPCORN POPULATIONS ABSTRACT-The popcorn maize populations CMS-42 and CMS-43 were undergone to the second selection cycle among and within half-sib families. The objectives were to increase the favorable allele frequencies and to estimate the genetics and phenotypics parameters from some agronomic traits. It was used an 14 x 14 simple lattice field design with 196 treatments. The former selection cycles were used as intercalar check. Plots consisted of one 5 m row with a plant density of 50,000 pilha. Although more importance was given to popping expansion, the mean values from the selected families showed the possibility to obtain high productivity with these populations. CMS-43 showed greater variability than CMS-42 as could be seen by the greater additive genetic variance component in CMS-43 than in CMS-42. The mean values for popping capacity were 14.62%, 15.21% and 19.65% for the checks, families and selected progenies, respectively with the population CMS-42, and 16.07 0%, 16.85% and 22.31%, respectiveiy, for CMS-43. For both populations, the narrow sense heritability for popping expansion was 60% and the estimated gain with selection among families was 16.80% for CMS-42 and 19.12% for CMS-43.
Full-text available
Article
Seis variedades de milho pipoca (Amarela, Roxa, Branca, Rosa-claro, Beija-flor e Viçosa), Zea mays L. (Poaceae), seus 15 híbridos F1’s e 15 híbridos F1’s recíprocos foram avaliados em um delineamento em blocos ao acaso, com três repetições, instalado em Viçosa, nos anos agrícolas 90/91 e 91/92, e Visconde do Rio Branco, em 91/92. Avaliaram-se os seguintes caracteres: altura da planta, altura da espiga, peso de grãos e capacidade de expansão. A variedade Viçosa foi indicada para ser usada em híbridos ou como fonte de linhagens para uso em programas de melhoramento intra ou interpopulacionais. O híbrido Viçosa x Roxa mostrou-se o mais desejável considerando rendimento e capacidade de expansão. As combinações Viçosa com Roxa e Rosa-claro com Beija-flor foram indicadas para iniciar programa de seleção recorrente recíproca, uma vez que foram as que mais se complementaram em relação ao peso de grãos e capacidade de expansão. Nestes programas, atenção especial deve ser dada à capacidade de expansão, que é a característica mais limitante nestes híbridos. Apesar de haver evidência de efeitos recíprocos para todos os caracteres, verificou-se pouca importância dos mesmos
Full-text available
Article
O objetivo deste trabalho foi gerar informações sobre a capacidade combinatória de dez genitores de milho-pipoca, em cruzamentos no esquema de dialelo circulante em dois ambientes, para futura utilização em programa de melhoramento. Os híbridos, genitores e testemunhas foram avaliados quanto à altura de plantas (AP), produção de grãos (PG) e capacidade de expansão (CE), em delineamento em blocos ao acaso, com quatro repetições. A análise conjunta do dialelo revelou reduzido comportamento diferencial dos genótipos, com predominância dos efeitos aditivos apenas para CE. Os genitores UNB2U-C1, BRS Angela, UNB2U-C2 e Viçosa-UFV foram os indicados para melhoramento intrapopulacional, ao passo que os híbridos de interesse para seleção interpopulacional foram PR-Ervália x BRS Angela e UNB2U-C1 x BRS Angela.
Full-text available
Article
Eight popcorn populations (CMS-42, CMS-43, Zelia, RS-20, Catedral, Zaeli, UEM-J1 and UEM-M2) were evaluated aiming at obtaintion of estimates of inbreeding depression and average genetic components and to compare these estimates to each genetic group. Each trial was carried out with plots in randomized complete block design with three replications in split-plot experiment, in which the plots were the inbreeding levels (S1 and S0) and in the sub-plots the populations in two years (1998/99 and 1999/00). The characteristics grain yield and popping expansion were evaluated. The GARDNER methodology (1965) was used for obtaining the estimates of average genetic components. It was observed lower values of inbreeding depression and predominance of additive effects for popping expansion in relation to yield. The values of inbreeding depression for grain yield were higher in the composite than in bath, improved and not improved varieties. Using these populations the probability of success in the obtaining vigorous and high popping expansion inbred lines is low.
Full-text available
Article
O objetivo foi avaliar, por simulação de dados, a eficiência de dialelos circulantes em relação aos completos, nas estimativas de capacidade geral de combinação (CGC) e capacidade específica de combinação (CEC). Foram simuladas 50 linhagens com 100 genes com distribuição independente, efeitos iguais, sem epistasia. Consideraram-se herdabilidades de 10%, 20%, 50% e 75%, em modelos com interação alélica aditiva e dominância completa. Utilizaram-se também dados de experimentos disponíveis na literatura, ambos comparados por meio de: correlação entre estimativas de CGC e CEC, e proporções de estimativas nos circulantes que ocorreram nos intervalos de confiança dos completos. Os resultados permitiram concluir que os dialelos circulantes são tão eficientes quanto os completos, tanto na classificação dos pais quanto à CGC e à CEC, como na magnitude desses parâmetros, e que o número de cruzamentos (s) afeta as estimativas de CGC e de CEC, embora com um valor pequeno de s seja possível obter boa concordância com as estimativas obtidas nos dialelos completos; com baixa herdabilidade, é vantajoso aumentar o número de cruzamentos de cada pai, embora com a metade do número de pais envolvidos já se consiga alta eficiência dos dialelos circulantes.
Full-text available
Article
Realizaram-se cruzamentos dialélicos entre nove populações de milho-pipoca (Zea mays L.) - UNB-2, RS-20, UEM-J1, UEM-M2, CMS-42, CMS-43, Viçosa, Beija-flor e a geração avançada do híbrido triplo Zélia - para a obtenção dos 36 híbridos F1's a fim de alcançar informação dos valores da capacidade combinatória e heterose. Essas populações e os F1's foram avaliados em ensaios, em dois anos agrícolas (98/99 e 99/00), em Iguatemi (PR), com três repetições por ano. Os resultados permitiram recomendar compostos formados entre as populações de alta capacidade geral de combinação (CGC) para rendimento de grãos, UNB-2 e CMS-42 e as populações de alta CGC para capacidade de expansão, Zélia, RS-20 e UEM-M2. As populações Beija-flor e RS-20 foram recomendadas para iniciar programa de seleção recorrente recíproca, pois se complementam em relação ao rendimento de grãos e à capacidade de expansão (CE). Para CE, houve heteroses positiva e negativa, verificando-se dominância não unidirecional.